Analysis of Water Quality Analyses in of a Constructed Treatment Wetland Designed forto Reduce NutrientsTreatment from in Everglades Agricultural Area Runoff
Binhe Gu, Jana Newman, Martha Nungesser and Micheal Michael J. Chimney
South Florida Water Management District, West Palm Beach, FloridaFL
The Everglades Nutrient Removal Project (ENRP) was a 3,6003,819 acre (ac1,545 ha) (treatment wetland built on former agricultural land in Palm Beach County, Florida (2638’N and 8025’W) and was operated by the District as a prototype stormwater treatment area (STA). was Phase 1 of STA-1W, serving as a prototype stormwater treatment area (STA). This wetlandThe ENRP was composed of four large treatment cells arranged as two parallel flow-ways; water with inflowwater movingedseparately from Cell 1 to Cell 3 (eastern flow-way) and from Cell 2 to Cell 4 (western flow-way). . The ENRP began began operation in August 1994,treating water in August 1994 and continued operation when it was incorporated into the footprint of a larger STA, STA-1 West in June 1999. and continued operation during the Phase 2 expansion that added on Cell 5, nearly doubling the footprint of the STA. This abstract willWe presentpresent the results of water quality analyses from based on nearly seven years of continuous operation of Cells 1- – 4,extending from A(August 1994 to April 2002).
The average hydraulic inflow to the ENRP rate was 195 cubic feet per second (cfs) and ranged from 7 to 407 cfs. . Water depth varied from 0.88 to 2.89 ft with an average of 1.95 ft. Nominal hydraulic retention times ranged from 9 to 99 days with an average of 25 days. . The vegetation community in Treatment Cells 1, 2toand 3 were was dominated by cattails (Typha latifolia and T. domingensis). . Treatment Cell 4 has beenwas actively maintained as a submerged aquatic vegetation (SAV) system community dominated by southern naiad (Najas quadalupensis), with lesser quantities of coontail (Ceratophyllum demersum) and pondweed (Potamogeton sp.).
Analysis of Wwater quality analyses wereas based on monthly or biweekly composite and grab samples collected from the ENRP’s cell inflow,and outflow,outflow and several interior stations. . Results indicate that dissolved iIon concentrations in at the ENRP the inflow waters exceeded the averages for North America freshwaters by several folds. . Calcium (Ca), alkalinity, iron (Fe), aluminum (Al), dissolved oxygen (DO), total suspended solids (TSS), turbidity, sulfate (SO4-2), total Pphosphorus (TP), soluble reactive Pphosphorus (SRP), total nitrogen (TN) and chlorophyll a (Chl a) decreased as water moved through the wetland. . Total organic Ccarbon (TOC), dissolved organic Ccarbon (DOC), conservative ions, conductivity and pH in the outflow were effectively relatively unchanged over the sampling period. . Secchi disk depth iwas the only variable parameter where average outflow values were significantly greater than at the inflow.
Inflow Aluminum and iron Fe concentrations, averaged 295 and 280 µg/L at inflows, respectively, and were reduced over 90% to 22 and 17 µg/L at the outflow, with over 90% removal by the treatment wetland. . Several pPossible sinks for iron Fe and Al include plant uptake, physical settling and co-precipitation with anions such as hydroxide and phosphate. . Removal of calcium Ca was also significant,significant; an average of 11 mg/L of Ca was retained by the wetland. . Two of the major sinks for calcium Ca are precipitation to sediments and encrustation on to vegetation such as periphyton and SAVsubmerged aquatic vegetation(SAV) as CaCO3. . Co-precipitation with Ca was thought to be responsible for some SRP removedal from the ENRP water column. . The Aaverage inflow TP concentration of 118 µg/L was reduced by 81% to 23 µg/L, and SRP was reduced from 57 µg/L to 9 µg/L. Uptake by higher plantsSAV and the associated periphyton, gravity settlementing and co-precipitation with CaCO3, iron Fe and Al are believed to have play been important roles mechanisms for for P removal.
The Aaverage TSS concentration of 6.9 mg/L was reduced to 2.1 mg/L,which and may partially explain account for the improvement ofin water clarity as well as some of the reduction in TP concentration. . The average water column Chlorophyll a concentration decreased from 17.6 µg/L to 3.5 µg/L. This dramatic decrease in planktonic Chl a and was likely attributed due to severe nutrient competition by higher plants which dominated the flora plant community. . Trend analysis indicates that Chl a continued to decrease while Secchi depth increased at the outflow over time.
Data analyses for the internal stations indicate varying levels of concentration reduction among the treatment cells. . Both front-endupper cells in each flow-way (Cells 1 and 2) removed TP as well as ironFe, nitrate (NO3-), Chl a, TSS, Al, and sulfate SO4 effectively. . Cells 1 and 2 both also had lower lower DO values at levels at theeir outflows relative to the inflows, while the back-end cells (Cells 3 and 4) of each flow-way Cells 3 and 4 increased the DO in the overlying water column. . Possible explanations for low increased oxygen DO consumptionlevelsof the front-end in cCells 1 and 2 include high nutrient loading of nutrients, Chl a and TSS which would lead to large amountshighof DO consumption during mineralization of organic matter mineralization. . Additionally, high turbidity may negatively affect photosynthesis by any SAV and periphyton communities situated in the front-endupper cells.
Cell 4 has proven to be more effective than Cell 3 in reducing TP and Ca concentrations in addition to having the highest P sediment accretion rates of the four cells. . Furthermore, Cell 4 had a higher rate of Ca concentration reduction compared to Cell 2. . It is probable that the SAV and periphyton biomass, which dominate the wetland, raised water column pH due to their higher photosynthetic rate, which further enhances CaCO3-P co-precipitation. . This would additionally account for the increased Ca removal noted in the back-end system of the western flow-way. .
In summary, the ENRP treatment wetland displays hadachieved high rates of concentration reduction for of nutrients and suspended particles. . Differences in removal efficiency among treatment cells are was ere due attributed to their positioning location in in the treatment traineach flow-way, surface area,and plant type community composition and coverage. . In order tTo further optimize the P removal in the STAs to enhance P removal, increased understanding of cycling pathways for major elements within the treatment wetlands is needed.
Dr. Binhe Gu,
Lower East Coast Division, MS-4830,
South Florida Water MManagement District,
3301 Gun Club Road
, West Palm Beach, FL 33406
Phone: 561-682-2556, Fax: 561-682-0100, ,